Surgical Management of Ovarian Disease in Infants,Children,and Adolescents: A 15-Year Review

Introduction

Ovarian pathology in infants and children, although a relatively less common occurrence than that in adult women, should be included in the differential diagnosis of all girls who present with abdominal pain, mass, or precocious puberty. Ovarian disease remains a rare indication for surgery in girls less than 20 years of age, and since most of these lesions are benign, ovarian preserving operations should be performed whenever feasible. ¹ Removal of ovarian pathology may be accomplished either by open laparotomy or less-invasive laparoscopic approaches. Laparoscopy has proved to be a major advancement in the management of ovarian disease due to the better visualization, reduced invasiveness, and shorter hospitalization. Ovarian procedures represent one of the most commonly performed laparoscopic procedures in routine adult practices. ²

Studies comparing open versus laparoscopic methods of managing ovarian pathology in adult women have demonstrated significant reductions in length of hospital stay, surgical complications, and hospital costs when ovarian lesions were excised by using a laparoscopic approach. ³ Despite the reported efficacy and presumed benefits of a laparoscopic approach in adults, evidence in support of this practice for ovarian diseases across all pediatric age groups is lacking. The aim of our study was to review our institutional experience with the surgical management of ovarian disease over the last 15 years to evaluate outcomes, complications, and potential benefits of the laparoscopic approach, compared to open methods, for benign ovarian lesions. We hypothesized that there has been a shift in the management strategy for ovarian pathology, which has trended from traditional open to minimally invasive approaches during this time period.

Methods

This was an institutional review board (IRB)-approved retrospective review of the Children's Hospital of Pittsburgh of UPMC (Pittsburgh, PA) medical records for 231 patients who underwent surgical management of ovarian pathology from January 1, 1992 to July 10, 2007. Patients undergoing surgery for suspected ectopic pregnancy, pelvic pain for known pelvic inflammatory disease, or concomitant illness requiring primary operation at the time of management of an ovarian lesion (i.e., incidental ovarian finding) were excluded. Demographics, clinical signs and symptoms, diagnostic methods, past medical and surgical history, operative time, length of hospital stay, and blood-loss data were obtained for each patient. Excluded from the surgical outcomes analyses were patients with malignant ovarian tumors (n = 10) who required chemotherapy as part of their initial hospital visit. Results were analyzed by comparison of means, Student's t-test, chi-squared test, and multivariate analysis, where indicated.

Results

From January 1, 1992 to July 10, 2007, we identified 231 patients that underwent surgery for the management of ovarian lesions (see Table 1). Of these patients, 73 (33%) were managed by using an open laparotomy method; 148 (67%) patients underwent minimally invasive surgery (MIS) for surgical management. All 10 patients with ovarian malignancies were managed by laparotomy. The mean age of the patients in this study was 14 years (range, 3 weeks to 20 years). The mean age for the patients in the laparotomy group versus the MIS group was 13 (range, 4 weeks to 19 years) and 14 years (range, 3 weeks to 20 years), respectively. In total, 90% patients were of Caucasian origin and 10% of patients were African American.

Preoperative signs and symptoms included abdominal pain (82%), nausea or vomiting (24%), palpable abdominal mass (10%), tenderness to palpation, rebound, or guarding (10%), fever (7%), abdominal distension (7%), constipation (4%), urinary frequency (3%), abnormal vaginal bleeding (1%), breast enlargement (n = 1), and hirsutism (n = 1). The following diagnoses (see Table 2) were determined on surgical exploration and pathologic examination: simple cyst (48%; n = 114), hemorrhagic cyst (19%; n = 45), mature teratoma (19%; n = 45), mucinous cystadenoma (4%; n = 9), dysgerminoma (3%; n = 6), immature teratoma (2%; n = 5), paratubal cyst (2%; n = 4), granulosa cell tumor (1%; n = 3), germ cell tumor (1%; n = 3), and serous cystadenoma (1%; n = 2). Benign pathology was described in the vast majority of cases, and there were patients who had more than one diagnosis recorded per specimen.

The ovarian lesions were grouped into three categories, based on their histologic composition: cystic, heterogenous with both solid and cystic components, and solid. Of the 231 lesions described, 181 (78%) were cystic, 35 (15%) were heterogenous, and 15 (7%) were solid. In the overall series, 13% of patients had prior abdominal procedures. Of these patients, 84% had their new ovarian diagnoses managed via open laparotomy, whereas only 16% of lesions were managed via laparoscopic methods. Of the patients that had available ultrasound or computed tomography (CT) scan reports, the overall median largest diameter was 6.5 cm (range, 2–29). Of patients in the laparoscopy group, the median largest diameter of the lesion was 6 cm (range, 2–26). In the laparotomy group, the median largest diameter was 11.5 cm (range, 6–29). In total, 39 patients (17%) were found to have torsion, as defined by intraoperative evaluation of blood flow or evidence of tissue necrosis.

Surgical outcomes, such as operative time, hospital length of stay, blood loss, and intraoperative fluid requirement, were evaluated only for patients with benign ovarian disease, since the standard of care in the management of malignant disease is open laparotomy and proper surgical staging (n = 221). The mean operative time for patients undergoing laparotomy for the removal of their benign ovarian lesion was 127 minutes (Table 3). Patients who underwent MIS had a mean operative time of 84 minutes. The mean length of stay for patients undergoing laparotomy versus laparoscopy was 4.54 and 1.85 days, respectively. Average blood loss, as reported by the anesthesiologist, was compared between the two groups. The mean blood loss in the laparoscopy group was 10.4 versus 71.4 cc in the laparotomy group. When a multivariate analysis was performed, there remained a significant difference, in favor of MIS, with respect to length of stay and blood loss (P = 0.001, P = 0.001). The mean intraoperative crystalloid requirement for the laparoscopy group was 859 versus 1226 cc in the laparotomy group. Only 1 patient, who underwent open salpingo-oopherectomy, omentectomy, and peritoneal biopsy for the removal of a granulosa cell tumor, required an intraoperative blood tranfusion.

There were a number of different operations performed for benign disease, which included aspiration, fenestration, cystectomy, and marsupialization. Other procedures included oopherectomy, salpingo-oopherectomy, detorsion, and oophoropexy. There were 2 intraoperative complications. In the laparoscopic group, 1 patient had severe bradycardia that required the use of pressors for stabilization. In the laparotomy group, 1 patient developed pressure sores, which was thought to be secondary to improper positioning. Four patients were readmitted for postoperative complications, 3 in the laparoscopic group. One patient had a wound infection, 1 patient had an umbilical hernia, and 1 developed contralateral ovarian torsion, which was thought to be secondary to surgical manipulation during a salpingo-oopherectomy. In the laparotomy group, there was 1 readmission due to postoperative fever. There was 1 unplanned conversion from laparoscopy to laparotomy for biopsy and open resection of a malignant tumor, which was unexpected by preoperative imaging. The mass was biopsied through a Pfannenstiel incision to avoid tumor spillage.

Discussion

We reviewed the surgical management of ovarian disease at our institution over a 15-year period. Most of the lesions were benign (221/231), with only a 4.5% incidence of malignant tumors. This is comparable to several pediatric series, where the incidence of malignant ovarian tumors ranged from 2 to 4.5%.^1,5,6 The operative approach (laparoscopic versus open) to ovarian pathology was dependent upon surgeon preference and preoperative imaging studies (i.e., ultrasound and CT scans). Whereas the majority of lesions were cystic (78%), making a laparoscopic approach to removal reasonable, nearly 20% of the solid lesions were excised via open biopsy and laparotomy due to the concern for upstaging a malignancy by tumor spillage, large size, and difficulty in removal by laparoscopic methods. In fact, deSilva et al. found that patients with lesions greater than 10 cm and associated palpable abdominal mass were more likely to have malignant tumors, although the incidence of cancer in their series of children was only 4.5%. ⁶ In our study, the decision to proceed with the laparaoscopic versus an open approach was based upon the confidence of the surgeon in removing the lesion without compromising oncologic principles. In 1 patient with an unexpected primarily solid ovarian lesion, conversion from laparoscopy to an open biopsy and excision was performed. The management of benign teratomas was performed predominantly by laparoscopy, if the lesion was primarily cystic. Drainage was performed percutaneously via an enlarged umbilical or Pfannenstiel incision, in most cases, if necessary. We had no reported cases of tumor spillage. The incidence of spillage of fluid in adult patients has been reported from 25 to 42.5%.^2,7 Ehrlich et al. described an effective method of combining a minimally invasive 5-cm Pfannenstiel incision and bag to remove fluid from large cystic masses prior to excision from the abdominal cavity. ⁸ They advocated this approach, since sometimes there is difficulty in differentiating benign from malignant tumors preoperatively, as over 50% of malignancies contain cystic components. It may be reasonable to assume that although spillage of cyst contents may be high, this should be avoided, since the risks to the patient can be significant. Poncelet et al. demonstrated that ovarian cystectomy in adults was associated with a high rate or risk of intraoperative cyst rupture and recurrence of borderline ovarian tumors (up to 30%), compared to patients undergoing unilateral salpingo-oophorectomy or bilateral salpingo-oophorectomy. ⁹

With respect to ovarian torsion, some studies in children have shown up to 42% of cases have ovarian pathology contributing to the torsion. ⁴ Other studies have instead advocated the use of ultrasonography to determine flow as an assessment of viability when concerned about acute ovarian torsion. ¹⁰ In our study, 39 of our patients (17%) were found to have torsion, but we found no correlation between the size of the lesion on ultrasound and the risk of torsion. For benign ovarian pathology, we found a significant difference in favor of the laparoscopic, compared to the open, approach with respect to length of stay, operative time, and blood loss. Yuen et al. showed that, when compared prospectively in adults with ovarian disease, laparoscopy provided a significant reduction in morbidity, length of hospital stay, postoperative analgesia requirement, and quicker recovery. ¹¹

No pediatric study has compared open to laparoscopic approaches for ovarian disease at this time. The trend in management of benign ovarian disease at our institution has clearly shifted toward MIS in the mid-1990s to the present day. The safety of the laparoscopic approach has been well documented in adults, children, and neonates.^7,12,13 Hildebaugh et al. compared laparoscopic versus open methods for treating adnexal masses in adults and found more complications in the open than in the laparoscopic cases. ³ Some have recently advocated a laparoscopic-assisted transumbilical extracorporeal ovarian cystectomy as an adjunct to the management of neonatal ovarian cysts. ¹⁴ The patients in their study had quicker operations, feed advancement, and earlier discharges, compared with similar patients undergoing a purely laparoscopic approach. Laparoscopy is currently the accepted approach to presumed benign adnexal masses in children and adolescents and is considered, by some, to be the “standard of care.” ¹⁵

Conclusions

In this 15-year review, there was a notable shift in favor of MIS in the management of benign ovarian disease. This approach was associated with shorter hospital stay, less operative blood loss, and shorter operative times, when compared to an open approach. When clinically indicated, laparoscopic surgery should be performed for presumed benign ovarian disease across all pediatric age groups.